D-dimer testing in the emergency department: age adjustment, inappropriate use, and ability to predict the extension and severity of pulmonary embolism

Objectives. To evaluate whether using D-dimer test results adjusted for age according to the formula proposed by Douma et al. improves diagnostic accuracy; to assess the appropriateness of ordering D-dimer tests on clinical suspicion of pulmonary embolism; and to explore the association of test results with the extension and severity of the embolism. Methods. Retrospective observational study of 1833 cases in which D-dimer testing was ordered for patients in our hospital's emergency department in the course of a year. We calculated sensitivity, specificity, and positive and negative predictive values using our hospital's D-dimer cutoff of 250 mu g/mL adjusted for age with a modification of Douma et al.'s formula. When information about pulmonary embolism extension and severity was on record, we assessed the correlation with test results. Results. Adjusting D-dimer level for age increased the number of true negatives and the specificity and positive predictive value of the test. D-dimer level correlated significantly with the extension of pulmonary embolism (r=0.41, P<.05) but not with clinical severity. Conclusions. Adjusting the D-dimer test result by age improves accuracy in the diagnosis of pulmonary embolism, even though clinical suspicion in Spain does not follow guideline recommendations. Our findings suggest that D-dimer level correlates with the extension but not the severity of pulmonary embolism.